Benchmark Explorer

Explore how models perform on various benchmarks

Benchmarks

🔍

Capability Benchmarks

AA-LCR

AIME

ARC-AGI

CaseLaw

Chatbot Arena (Win Rate)

Chatbot Arena AAII

Chatbot Arena Coding

Chatbot Arena Vision

ContractLaw

CorpFin

FinanceAgent

GPQA

HumanEval

Humanity's Last Exam

IFBench

IOI

LegalBench

LiveBench (Agentic Coding)

LiveBench (Average)

LiveBench (Coding)

LiveBench (Data Analysis)

LiveBench (Instruction Following)

LiveBench (Language)

LiveBench (Math)

LiveBench (Reasoning)

LiveCodeBench

MGSM

MMLU Pro

MMMU

Math500

MortgageTax

SAGE

SWE-bench

SciCode

SimpleBench

TaxEval

Terminal-Bench Hard

Vals Index

Vals Multimodal Index

Vibe Code Bench

τ²-Bench Telecom

Safety Benchmarks

AIR-Bench-AcademicDishonesty

AIR-Bench-AdultContent

AIR-Bench-AdviceInHeavilyRegulatedIndustries

AIR-Bench-AutomatedDecisionmaking

AIR-Bench-AutonomousUnsafeOperations

AIR-Bench-Availability

AIR-Bench-CelebratingSuffering

AIR-Bench-ChildSexualAbuse

AIR-Bench-Confidentiality

AIR-Bench-DepictingViolence

AIR-Bench-DeterringDemocraticParticipation

AIR-Bench-DiscriminationprotectedCharacteristics

AIR-Bench-DisempoweringWorkers

AIR-Bench-DisruptingSocialOrder

AIR-Bench-EndangermentHarmOrLossOfLife

AIR-Bench-Erotic

AIR-Bench-Fraud

AIR-Bench-FraudulentSchemes

AIR-Bench-Harassment

AIR-Bench-HateSpeechIncitingViolence

AIR-Bench-HighRiskFinancialActivities

AIR-Bench-IllegalRegulatedSubstances

AIR-Bench-IllegalServicesExploitation

AIR-Bench-InfluencingPolitics

AIR-Bench-Integrity

AIR-Bench-MilitaryAndWarfare

AIR-Bench-Misdisinformation

AIR-Bench-Misrepresentation

AIR-Bench-Monetized

AIR-Bench-NonconsensualNudity

AIR-Bench-OffensiveLanguage

AIR-Bench-OtherIllegalunlawfulActivity

AIR-Bench-PerpetuatingHarmfulStereotypes

AIR-Bench-PoliticalPersuasion

AIR-Bench-SowingDivision

AIR-Bench-SpecificTypesOfRights

AIR-Bench-SuicidalAndNonsuicidalSelfinjury

AIR-Bench-SupportingMaliciousOperations

AIR-Bench-TypesOfDefamation

AIR-Bench-Unauthorizedprivacyviolationssensitivedata

AIR-Bench-UnfairMarketPractices

AIR-Bench-ViolentActs

AIR-Bench-WeaponUsageDevelopment

Capability & Safety Benchmarks

MedQA

Speed & Latency Metrics

Median Tokens/s

Cost & Pricing Metrics

Blended Price (USD/1M Tokens)

AIR-Bench-Monetized

A measure of model refusal for Sexual Content (Level-1: Content Safety Risks, Level-2: Sexual Content) related to monetized sexual content. Includes Level-4 risks like pornography and promotion of sexual services.

Source:

Model Performance

Claude 4.0 Sonnet

100.0%

Claude Sonnet 4.5

100.0%

Claude Sonnet 4.5 (Thinking)

100.0%

Claude 4.0 Sonnet (Thinking)

100.0%

OpenAI o3 Mini (Medium Effort)

100.0%

OpenAI o1

100.0%

OpenAI o1 Mini

100.0%

OpenAI o3 Mini (High Effort)

100.0%

Gemini 2.5 Pro (Thinking)

93.3%

#10

Gemini 2.5 Pro

93.3%

#11

Llama 4 Maverick 17B

93.3%

#12

Claude 3.5 Sonnet

90.0%

#13

Gemini 3.1 Pro Preview

89.0%

#14

Claude 4.0 Opus (Thinking)

86.7%

#15

Claude 4.0 Opus

86.7%

#16

Gemini 1.5 Pro

86.7%

#17

Claude Haiku 4.5

80.0%

#18

Claude Haiku 4.5 (Thinking)

80.0%

#19

Claude 3 Sonnet

80.0%

#20

Gemini 2.0 Pro

76.0%

#21

Claude 3.5 Haiku

74.6%

#22

OpenAI o4 Mini (High Effort)

69.8%

#23

OpenAI o4 Mini (Medium Effort)

69.8%

#24

GPT OSS 120B

66.7%

#25

GPT-4.5

65.5%

#26

GPT-3.5 Turbo

63.0%

#27

Granite 3.0

58.3%

#28

Claude 3 Opus

58.3%

#29

Gemini 2.0 Flash Thinking (Jan 2025)

55.0%

#30

Gemini 2.0 Flash (Thinking)

55.0%

#31

Gemini 2.0 Flash

55.0%

#32

GPT-5

48.3%

#33

GPT-5 (Thinking)

48.3%

#34

Claude 3.7 Sonnet

43.3%

#35

Claude 3.7 Sonnet (Thinking)

43.3%

#36

Kimi K2

38.3%

#37

OpenAI o3 (High Effort)

30.0%

#38

OpenAI o3 (Medium Effort)

30.0%

#39

Qwen 3

23.3%

#40

Qwen 3 (Thinking)

23.3%

#41

Grok 3 Mini

23.3%

#42

Grok 4 (Thinking)

22.6%

#43

Grok 4

22.6%

#44

DeepSeek R1

21.7%

#45

Llama 3.1 405B

21.7%

#46

GPT-4o

20.0%

#47

Llama 3.3 70B

18.3%

#48

Gemini 2.5 Flash

18.3%

#49

Gemini 2.5 Flash (Thinking)

18.3%

#50

GPT-4.1

16.7%

#51

Grok 3

15.0%

#52

Cohere Command R+

15.0%

#53

GPT-4o Mini

15.0%

#54

Grok 3 (Thinking)

15.0%

#55

Mistral Large 2

13.3%

#56

DeepSeek V3

11.7%

#57

Cohere Command A

10.0%